CN110414022A - A kind of method for early warning and system of wind generator set blade cracking - Google Patents
A kind of method for early warning and system of wind generator set blade cracking Download PDFInfo
- Publication number
- CN110414022A CN110414022A CN201810393681.0A CN201810393681A CN110414022A CN 110414022 A CN110414022 A CN 110414022A CN 201810393681 A CN201810393681 A CN 201810393681A CN 110414022 A CN110414022 A CN 110414022A
- Authority
- CN
- China
- Prior art keywords
- value
- data
- rate
- temperature change
- early
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F03—MACHINES OR ENGINES FOR LIQUIDS; WIND, SPRING, OR WEIGHT MOTORS; PRODUCING MECHANICAL POWER OR A REACTIVE PROPULSIVE THRUST, NOT OTHERWISE PROVIDED FOR
- F03D—WIND MOTORS
- F03D17/00—Monitoring or testing of wind motors, e.g. diagnostics
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/70—Wind energy
- Y02E10/72—Wind turbines with rotation axis in wind direction
Landscapes
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Sustainable Development (AREA)
- Sustainable Energy (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Wind Motors (AREA)
Abstract
The present invention provides the method for early warning and system of a kind of wind generator set blade cracking, the described method comprises the following steps: obtaining the actual operating data of wind power generating set；The derivative variable of early warning relevant to wind generator set blade cracking is calculated according to the actual operating data of acquisition；Data prediction is carried out to the derivative variable of the early warning of the actual operating data of acquisition and calculating respectively；Early-warning Model is constructed based on pretreated actual operating data and the derivative variable of early warning, and is cracked according to the Early-warning Model of building to blade and carries out early warning.The present invention can overcome the defect that the unconspicuous problem of temperature change of part MW class wind turbine turbines vane cracking and the detection of pitch motor sensor, provides the real time on-line monitoring of blade operation and gives warning in advance, provides guarantee for the reliable and stable of blade.
Description
Technical field
The present invention relates to technical field of wind power, in particular, being related to a kind of pre- police of wind generator set blade cracking
Method and system.
Background technique
Three blade appearances of wind power generating set (hereinafter referred to as " unit ") are due to weather, environment and itself material etc.
The case where factor is possible to crack after running for a period of time, causes blade cracking, fracture.Wherein, leaf destruction is more
Occur in root of blade and middle part of blade, in the form of fractureing, blade cracking mostly occurs in blade tip and middle part of blade edge, in vertical
It dehisces form to separation.The cracking of blade and fracture will cause heavy damage to unit operation, therefore, the research of early warning technology and
Exploration is a core technology of wind power generating set industry.Main group of impeller pitch-controlled system become pod, pitch variable bearings,
Pitch control cabinet, pitch motor, wheel hub and variable pitch toothed belt etc., wherein the operation logic of wheel hub is that blade absorbs wind energy, is driven
Hub rotation, if there is crackle in blade operation, the impeller unbalanced load of unit increases, and corresponding pitch motor just needs more
Big torque current carrys out the movement of driving blade variable pitch, therefore, immesurable blade cracking, fracture characteristics is mapped to measurable change
On paddle Motor torque, pitch motor operation torque is bigger, and the temperature change of pitch motor is more violent.
Mainly there are artificial detection, unmanned plane detection and big data pre- online the blade appearance Detection Techniques of unit at present
The methods of police-spy's survey.Wherein, detection is carried out to blade appearance by big data on-line early warning detection method and is mainly based upon load
Variation, will test unit load change comparison operate normally unit load change, if detection unit blade loading get over
Greatly, then show that blade has the probability of crackle bigger.Big data on-line early warning detection method compares personal monitoring and unmanned plane image
Identification has the effect of more succinct, stable, safety and saves economic value, but part MW class wind turbine group blade
The sensor detection temperature change of pitch motor is not obvious when cracking, it is thus impossible to carry out in advance or timely that failure is pre-
It is alert, the blade cracking of part MW class wind turbine group is just caused, unit operation troubles is caused.
Summary of the invention
The present invention aiming at the shortcomings of the prior art, provides a kind of method for early warning of wind generator set blade cracking and is
System.
An aspect of of the present present invention provides a kind of method for early warning of wind generator set blade cracking, and the method includes following
Step: the actual operating data of wind power generating set is obtained；According to the calculating of the actual operating data of acquisition and wind power generating set
The derivative variable of the relevant early warning of blade cracking；The derivative variable of the early warning of the actual operating data of acquisition and calculating is counted respectively
Data preprocess；Early-warning Model is constructed based on pretreated actual operating data and the derivative variable of early warning, and according to the pre- of building
Alert model, which cracks to blade, carries out early warning.
Preferably, the actual operating data includes the acquisition time of actual operating data, the temperature of pitch motor and leaf
The propeller pitch angle of piece.
Preferably, described that early warning relevant to wind generator set blade cracking is calculated according to the actual operating data of acquisition
The step of derivative variable includes: the time difference for calculating the acquisition time of the actual operating data, wherein the time difference is to obtain
The latter time of adjacent time in the time is taken to subtract the previous time；The temperature of pitch motor is accordingly calculated according to the time difference
Degree difference and pitch angular difference；The rate of temperature change of pitch motor is calculated according to the temperature difference of pitch motor and time difference.
Preferably, described that data prediction is carried out to the derivative variable of the early warning of the actual operating data of acquisition and calculating respectively
The step of include: missing values and null value in the removal actual operating data；Remove the actual operating data and described pre-
Error information in alert derivative variable, wherein the error information includes that the temperature of pitch motor is more than or equal to 150 DEG C of data
It is greater than 50 ° of data with pitch angular difference.
Preferably, described based on pretreated actual operating data and the derivative variable building Early-warning Model of early warning and basis
The step of Early-warning Model of building carries out early warning to blade cracking includes: to construct early warning mould according to the rate of temperature change of pitch motor
Type；The rate of temperature change accounting value of the wind generator set blade for early warning is sought according to the Early-warning Model of building, and when temperature
Degree change rate accounting value triggers alarm when being more than or equal to the alarm value G for triggering blade cracking, wherein the rate of temperature change
Accounting value be meet the rate of temperature change of specified conditions quantity account for pitch motor rate of temperature change total amount ratio.
Preferably, the step of data building Early-warning Model of the rate of temperature change according to pitch motor includes: to establish
Assignment matrix corresponding with rate of temperature change, the initial value of the element in the assignment matrix are 0；In temperature change rate
The first value is assigned a value of in the data in preset range, and by element corresponding in assignment matrix；In temperature change rate
Adjacent and symbol in the preset range is opposite and/or is separated by and any two data that symbol is opposite, and by assignment
Corresponding element value is assigned a value of second value in matrix；Meet the data of the following conditions in temperature change rate and by assignment
Corresponding element value is assigned a value of third value in matrix: adjacent and symbol is opposite and/or is separated by and symbol is opposite any two
A rate of temperature change, one in the absolute value of any two rate of temperature change is more than or equal to threshold alpha, another is less than threshold
Value α, and the absolute value of the difference of any two rate of temperature change is greater than the data of threshold gamma；By to the member in assignment matrix
The assignment that element carries out the first value, second value and third value obtains new assignment matrix, to complete the building of Early-warning Model, wherein
The specified conditions are α < < | Δ S | < β, Δ S are rate of temperature change, α=0.11, β=0.657, the threshold gamma=
0.18, the alarm value G=0.01.
Preferably, the Early-warning Model according to building seeks the temperature change of the wind generator set blade for early warning
The step of rate accounting value includes:
The new assignment is accounted for according to the quantity that new assignment matrix calculates remaining first value corresponding with each pitch motor
The ratio value of matrix total data line number, to obtain rate of temperature change accounting value corresponding with each pitch motor.
A kind of early warning system of wind generator set blade cracking, the system are provided according to another aspect of the present invention
Include: that actual operating data obtains module, is configured as obtaining the actual operating data of wind power generating set；Early warning derives variable
Computing module is configured as being spread out according to the actual operating data of acquisition calculating early warning relevant to wind generator set blade cracking
The amount of changing；Data preprocessing module, be configured to the derivative variable of the early warning to the actual operating data of acquisition and calculating into
Line number Data preprocess；Warning module is configured as pre- based on pretreated actual operating data and the derivative variable building of early warning
Alert model, and cracked according to the Early-warning Model of building to blade and carry out early warning.
Preferably, the actual operating data includes the acquisition time of actual operating data, the temperature of pitch motor and leaf
The propeller pitch angle of piece.
Preferably, the derivative variable computing module of the early warning is configured as: when calculating the acquisition of the actual operating data
Between time difference, wherein the time difference is that the latter time of adjacent time subtracts the previous time in acquisition time；According to described
Time difference accordingly calculates the temperature difference and pitch angular difference of pitch motor；It is calculated according to the temperature difference of pitch motor and time difference
The rate of temperature change of pitch motor.
Preferably, the data preprocessing module is configured as: removing the missing values and sky in the actual operating data
Value；Remove the error information in the actual operating data and the derivative variable of the early warning, wherein the error information includes becoming
Data of the temperature of paddle motor more than or equal to 150 DEG C and pitch angular difference are greater than 50 ° of data.
Preferably, the warning module is configured as: Early-warning Model establishes unit, according to the rate of temperature change of pitch motor
Construct Early-warning Model；Early warning judging unit seeks the wind generator set blade for early warning according to the Early-warning Model of building
Rate of temperature change accounting value, and the triggering when rate of temperature change accounting value is more than or equal to the alarm value G for triggering blade cracking
Alarm, wherein the rate of temperature change accounting value is to meet the quantity of the rate of temperature change of specified conditions to account for the temperature of pitch motor
Spend the ratio of the total amount of change rate.
Preferably, the Early-warning Model is established unit and is configured as: establishing assignment matrix corresponding with rate of temperature change, institute
The initial value for stating the element in assignment matrix is 0；Data in temperature change rate in the preset range, and by assignment square
Corresponding element is assigned a value of the first value in battle array；Adjacent and symbol in temperature change rate in the preset range
It is opposite and/or be separated by and any two data that symbol is opposite, and element value corresponding in assignment matrix is assigned a value of the
Two-value；Meet the data of the following conditions in temperature change rate and element value corresponding in assignment matrix is assigned a value of
Three values: adjacent and symbol is opposite and/or is separated by and any two rate of temperature change that symbol is opposite, any two temperature become
One in the absolute value of rate is more than or equal to threshold alpha, another is less than threshold alpha, and any two rate of temperature change it
Absolute value of the difference is greater than the data of threshold gamma；By carrying out the first value, second value and third value to the element in assignment matrix
Assignment obtains new assignment matrix, to complete the building of Early-warning Model, wherein the specified conditions are α < < | Δ S | < β, Δ
S is rate of temperature change, α=0.11, β=0.657, threshold gamma=0.18, the alarm value G=0.01.
Preferably, the early warning judging unit is also configured to be calculated and each pitch motor according to new assignment matrix
The quantity of corresponding remaining first value accounts for the ratio value of the new assignment matrix total data line number, to obtain and each pitch motor
Corresponding rate of temperature change accounting value.
Another aspect provides a kind of computer readable storage mediums, are stored with computer program, feature
It is, when the computer program is run by processor, the processor executes wind generator set blade as described above and opens
The method for early warning split.
Another aspect provides a kind of computer equipments, the storage including processor and storage computer program
Device, which is characterized in that when the computer program is run by processor, the processor executes wind-driven generator as described above
The method for early warning of group blade cracking.
In the present invention, the rate of temperature change based on pitch motor constructs Early-warning Model, and then realizes the pre- police conduct of big data
The problem of power generator group blade cracks realizes the real time on-line monitoring of blade operation and the work that gives warning in advance, and is blade
It is reliable and stable to provide guarantee.
Detailed description of the invention
Pass through the description carried out below in conjunction with attached drawing, above and other aspects, the feature of exemplary embodiment of the present invention
It will be more readily apparent from advantage, in the accompanying drawings:
Fig. 1 shows a kind of method for early warning flow chart of wind generator set blade cracking of embodiment according to the present invention；
Fig. 2 shows the flow charts of the building Early-warning Model of embodiment according to the present invention；
Fig. 3 shows the flow chart of the progress early warning judgement of embodiment according to the present invention；
Fig. 4 show exemplary embodiment according to the present invention by data set 1 and data set 2 to wind power generating set leaf
Piece cracking carries out the flow chart of early warning；
Fig. 5 shows a kind of early warning system block diagram of wind generator set blade cracking of embodiment according to the present invention；
Fig. 6 shows the block diagram of the warning module of embodiment according to the present invention.
Specific embodiment
The present invention of the description to help comprehensive understanding to be defined by the claims and their equivalents referring to the drawings is provided
Exemplary embodiment.Description referring to the drawings includes various specific details to help to understand, but the specific detail
It only is seen as illustrative.Therefore, it will be appreciated by those of ordinary skill in the art that not departing from scope and spirit of the present invention
In the case where, the embodiments described herein can be made various changes and modifications.In addition, for clarity and briefly, public affairs can be omitted
Know the description of function and structure.
Fig. 1 is the method for early warning process for showing a kind of wind generator set blade cracking of embodiment according to the present invention
Figure.
As shown in Figure 1, firstly, obtaining the actual operating data of wind power generating set in step S100.Specifically, it obtains
The actual operating data of wind power generating set include the acquisition time of actual operating data, the temperature of pitch motor and blade
Propeller pitch angle.According to an embodiment of the invention, due to wind power generating set is widely distributed, environmental difference is big, geographical location height not
The difference of the respective characteristic bring pitch motor temperature value of same and different units is also big, and therefore, here, selection is using same
Platform unit is tested, it is assumed that is acquired from SCADA (Supervisory Control And Data Acquisition) data
With one month actual operating data for obtaining the wind power generating set in supervisor control, wherein the actual motion of acquisition
Data include the pitch angular data A of time data t, the temperature data T of pitch motor and blade.
In step S200, early warning relevant to wind generator set blade cracking is calculated according to the actual operating data of acquisition
Derivative variable.Specifically, the time difference of the acquisition time of actual operating data is first calculated, wherein the time difference is in acquisition time
The latter time of adjacent time subtracts the previous time, then, is accordingly calculated and pitch motor according to the time difference of calculating
The rate of temperature change of pitch motor is calculated in temperature difference and pitch angular difference, temperature difference and time difference further according to pitch motor,
Thus it obtains and cracks the poor derivative variant time of relevant early warning, temperature difference of pitch motor, pitch to wind generator set blade
The rate of temperature change of angular difference and pitch motor.According to an embodiment of the invention, assuming according in the actual operating data of acquisition
Time data t, the temperature data T of pitch motor and pitch angular data A calculate separately the temperature difference for obtaining time difference, pitch motor
With pitch angular difference, wherein the calculation formula of time difference is as follows:
Δ t=tn-tn-1 (1)
Δ t is that the latter time of adjacent time in acquisition time subtracts previous time, t in above formulanFor n-th of time number
According to.
The calculation formula of the temperature difference of pitch motor is as follows:
Δ T=Tn-Tn-1 (2)
Δ T is the temperature difference of the temperature data corresponding with time data t in the actual operating data obtained in above formula,
TnFor the corresponding pitch motor temperature of n-th of time data.
The calculation formula of pitch angular difference is as follows:
Δ A=An-An-1 (3)
Δ A is the variable pitch of the pitch angular data corresponding with time data t in the actual operating data obtained in above formula
Differential seat angle, AnFor the corresponding pitch angular data of n-th of time data.
According to an embodiment of the invention, obtained time difference, pitch motor temperature difference and propeller pitch angle difference are not carried out head
Row mends 0, is incorporated to Data data matrix.Assuming that choosing continuous 6 time data from the actual operating data of acquisition and should be with
The temperature data and pitch angular data of the corresponding pitch motor of time data carry out time difference, variable pitch according to the data of selection
The calculating of motor temperature difference and pitch angular difference, first trip generate the data matrix of 6 rows 7 column after mending 0.Assuming that in the actual motion of acquisition
One group of time data is arbitrarily selected in data is 3,7,13,19,24,30, then according to the temperature data of corresponding pitch motor and
Poor to derivative variant time, pitch motor temperature difference and pitch angular difference calculate pitch angular data respectively, obtain a result as
Shown in following table:
The calculated result of the derivative variable of table 1
Blade is calculated according to the temperature difference of time difference shown in table 1 and the pitch motor of blade 1, blade 2 and blade 3
1, the rate of temperature change of the pitch motor of blade 2 and blade 3, calculation formula are as follows:
Δ S=Tn-Tn-1/tn-tn-1 (4)
According to the example above, the rate of temperature change of the pitch motor of corresponding three blades is calculated by formula (4),
Wherein, the rate of temperature change of the pitch motor of blade 1 are as follows: 0,0.11, -0.35,0.33, -0.38,0.17, the variable pitch electricity of blade 2
The rate of temperature change of machine are as follows: 0,0.29,0.17, -0.17, -0.08, -0.03, the rate of temperature change of the pitch motor of blade 3 are as follows:
0、0.10、-0.08、-0.33、0.2、-0.45。
In step S300, data are carried out to the derivative variable of the early warning of the actual operating data of acquisition and calculating respectively and are located in advance
Reason.Specifically, the missing values and null value in actual operating data are removed, then are removed in actual operating data and the derivative variable of early warning
Error information.According to an embodiment of the invention, assuming after analyzing the actual operating data of acquisition, the NA obtained (is lacked
Mistake value) and assigning null data in have discrete data, also have continuous data, then, for continuously be greater than 10 minutes NA (missing
Value) data and assigning null data can carry out the deletions of full line data, and for continuously less than 10 minutes NA (missing values) data and
Assigning null data and discrete data then carry out corresponding deletion.It should be understood that the act of the above-mentioned deletion for missing values and assigning null data
Example is only illustrative examples, and the method that the adoptable data of the present invention are deleted is without being limited thereto.Actual operating data and pre- is removed again
Error information in alert derivative variable, for example, the temperature of removal pitch motor is big more than or equal to 150 DEG C of data and pitch angular difference
In 50 ° of data, here, error information refers to the obvious abnormal data in wind power generating set operational process.On it should be understood that
Stating for the citing of error information is only illustrative examples, and the adoptable error information of the present invention is without being limited thereto.
In step S400, Early-warning Model, and root are constructed based on pretreated actual operating data and the derivative variable of early warning
It cracks according to the Early-warning Model of building to blade and carries out early warning.Specifically, it is constructed according to the data of the rate of temperature change of pitch motor
Then Early-warning Model is accounted for according to the rate of temperature change that the Early-warning Model of building seeks the wind generator set blade for early warning
Ratio, and alarm is triggered when rate of temperature change accounting value is more than or equal to the alarm value G for triggering blade cracking, wherein temperature
Degree change rate accounting value be meet the rate of temperature change of specified conditions quantity account for pitch motor rate of temperature change total amount
Ratio.The process of building Early-warning Model according to an embodiment of the present invention is described in detail below with reference to Fig. 2.
Fig. 2 is the flow chart for showing the building Early-warning Model of embodiment according to the present invention.
As shown in Fig. 2, in step s 201, establishing assignment matrix corresponding with rate of temperature change, the member in assignment matrix
The initial value of element is 0.Specifically, increase L1, L2, L3 tri- in data Data to arrange, and the initial value that L1, L2, L3 tri- is arranged is all provided with
It is 0, wherein L1, L2, L3 respectively represent that the rate of temperature change of pitch motor 1-3 of blade 1, blade 2 and blade 3 is corresponding to be reflected
Penetrate data.According to an embodiment of the invention, assume to take 8 rate of temperature change of each blade, establish respectively and three blades
The corresponding mapping of rate of temperature change, the then assignment matrix obtained are 0 matrix of 8 rows 3 column.
In step S202, data in temperature change rate in the preset range, and by assignment matrix with its
Corresponding element is assigned a value of the first value.Specifically, the rate of temperature change of the pitch motor in specified conditions is mapped to assignment square
0 value at battle array carries out assignment again.According to the example above, here, chooses L1 column data element and carry out element assignment, i.e., to blade
The rate of temperature change Δ S of 1 pitch motor is analyzed, it is assumed that the pitch motor of blade 1 corresponding with L1 column data element
Rate of temperature change Δ S be 0,0.10, -0.08, -0.33,0.2, -0.45, -0.6, -0.4, take specified conditions be α < < | Δ S
| < β, wherein α=0.11, β=0.657, then by 0.11 < < | and Δ S | the data position of < 0.657 is assigned a value of first
Value, it is 0,0,0,1,1,1,1,1 that the first value, which is set as 1 to get L1 column data element out, here.It should be understood that above-mentioned for specific
The citing of condition is only illustrative examples, and the adoptable specified conditions of the present invention are without being limited thereto.
In step S203, the adjacent and symbol in temperature change rate in preset range is opposite and/or is separated by
And any two data that symbol is opposite, and element value corresponding in assignment matrix is assigned a value of second value.According to above-mentioned
Citing, -0.33 and 0.2 is adjacent and opposite symbol data, and -0.45 and 0.2 is also adjacent and opposite symbol data, 0.2
It is to be separated by and opposite data with -0.6, then the assignment of corresponding position is changed to second value, it is assumed that second value 2, then L1
The corresponding assignment of column data element is changed to 0,0,0,2,2,2,2,1.
In step S204, the data of the following conditions are met in temperature change rate and will be corresponding in assignment matrix
Element value be assigned a value of third value: adjacent and symbol is opposite and/or is separated by and any two rate of temperature change that symbol is opposite, appoints
One to anticipate in the absolute value of two rate of temperature change is more than or equal to threshold alpha, another is less than threshold alpha, and any two temperature
The absolute value of the difference of change rate is greater than the data of threshold gamma.According to the example above, it is assumed that threshold gamma=0.18, then in blade 1
In the rate of temperature change 0,0.10, -0.08, -0.33,0.2, -0.45, -0.6, -0.4 of pitch motor, 0.10 and -0.33 is separated by
And symbol is greater than 0.11 on the contrary, simultaneously meeting in absolute value one simultaneously, between a condition and two data less than 0.11
Distance be therefore the assignment of corresponding position is changed to third value by 0.43 condition of value 0.18 greater than γ, it is assumed that the
Three values are 3, then the corresponding assignment of L1 column data element is changed to 0,3,0,3,2,2,2,1.It should be understood that above-mentioned for threshold gamma
Citing is only illustrative examples, and the adoptable threshold gamma of the present invention is without being limited thereto.
In step S205, obtained by the assignment for carrying out the first value, second value and third value to the element in assignment matrix
New assignment matrix out, to complete the building of Early-warning Model.According to the example above, pass through the temperature of the pitch motor to three blades
Degree change rate carries out analysis and assignment respectively to obtain new assignment matrix, wherein assignment 1 is expressed as wind caused by blade cracking
The pitch motor temperature anomaly of power generator group, assignment 2 and 3 item are expressed as pitch motor temperature anomaly caused by other reasons,
Such as the reasons such as card paddle, bearing breaking.Finally, being carried out according to the Early-warning Model of building to wind generator set blade cracking pre-
It is alert.
Fig. 1 is returned, the temperature change of the wind generator set blade for early warning is sought further according to the Early-warning Model of building
Rate accounting value, and alarm is triggered when rate of temperature change accounting value is more than or equal to the alarm value G for triggering blade cracking.Specifically
It is total to account for the new assignment matrix according to the quantity that new assignment matrix calculates remaining first value corresponding with each pitch motor for ground
Then the ratio value of number of data lines judges temperature change to obtain rate of temperature change accounting value corresponding with each pitch motor
The size of rate accounting value and alarm value G.As shown in figure 3, counting in new assignment matrix assignment in the column of L1, L2, L3 tri- respectively
The ratio of each column total amount is accounted for for 1 quantity, seeks the rate of temperature change accounting value P of the pitch motor of three blades1、P2And P3,
Further according to the rate of temperature change accounting value P of the pitch motor of three blades1、P2And P3, judge whether at least one temperature change
Rate accounting value is more than or equal to the alarm value G of blade cracking, if so, be judged as early warning unit and continue early warning judgement, it is no
Alarm is not triggered then.According to the example above, it is assumed that alarm value G=0.01, here, alarm value are by interior damage in recent years
It more units and is monitored analysis with the actual operating data of the unit operated normally in the period and obtains.By blade 1
The rate of temperature change accounting value P of pitch motor1=0.125 can determine whether the wind power generating set for triggering early warning unit, triggering
Alarm.It should be understood that the above-mentioned citing for alarm value is only illustrative examples, the adoptable alarm value of the present invention is without being limited thereto.
According to an embodiment of the invention, by one month actual operating data of the wind power generating set obtained to wind
After power generator group blade cracking carries out early warning, if being judged as triggering early warning unit, also need the wind power generating set
One month actual operating data is divided into first half moon data and later half moon data, and respectively to first half moon data and later half moon data
Carry out the early warning of Early-warning Model as described above.As shown in figure 4, data set 1 is first half moon data, data set 2 is later half moon data,
Then respectively to the early warning value G of data set 11With the early warning value G of data set 22Judged.Due to the reality in wind power generating set
A possibility that in operation, temperature change is more violent, and wind power generating set is damaged is bigger, so, in the early warning value G to data set 11
With the early warning value G of data set 22When being judged, compare G1And G2Size, and work as data set 1 early warning value G1Less than data set
2 early warning value G2When triggering early warning and sound an alarm, otherwise, do not trigger alarm, early warning terminates.
Fig. 5 is the early warning system block diagram for showing a kind of wind generator set blade cracking of embodiment according to the present invention.
Referring to Fig. 5, the early warning system 500 of wind generator set blade cracking may include that actual operating data obtains module
501, the derivative variable computing module 502 of early warning, data preprocessing module 503 and warning module 504.Implementation according to the present invention
The early warning system 500 of example, wind generator set blade cracking can be by various computing devices (for example, computer, server, work
Stand) Lai Shixian.Specifically, actual operating data obtains module 501 and is configured as obtaining the practical fortune of wind power generating set
Row data.The derivative variable computing module 502 of early warning is configured as according to the calculating of the actual operating data of acquisition and wind-driven generator
The derivative variable of the relevant early warning of group blade cracking.Data preprocessing module 503 is configured to the actual motion number to acquisition
Data prediction is carried out according to the derivative variable of early warning with calculating.Warning module 504 is configured as based on pretreated practical fortune
Row data and the derivative variable of early warning construct Early-warning Model, and are cracked according to the Early-warning Model of building to blade and carry out early warning.
According to an embodiment of the invention, it includes actual motion that actual operating data, which obtains actual operating data in module 501,
The propeller pitch angle of the acquisition time of data, the temperature of pitch motor and blade.The derivative variable computing module 502 of early warning is according to practical fortune
The actual operating data that row data acquisition module 501 obtains calculates the derivative change of early warning relevant to wind generator set blade cracking
Amount, specifically, the derivative variable computing module 502 of early warning first calculate the time difference of the acquisition time of actual operating data, wherein when
Between difference be that the latter time of adjacent time subtracts the previous time in acquisition time, accordingly calculated further according to the time difference of calculating
The temperature difference and pitch angular difference of pitch motor, finally, calculating the temperature of pitch motor according to the temperature difference of pitch motor and time difference
Spend change rate.
Data preprocessing module 503 obtains in module 501 and the derivative variable computing module 502 of early warning actual operating data
Data carry out data prediction, removed in the derivative variable computing module 502 of data preprocessing module 503 and early warning respectively
Missing values and null value and error information, wherein error information includes the temperature error data and pitch angular difference of pitch motor
Error information.It is greater than in the data and propeller pitch angle difference data in the temperature data of pitch motor more than or equal to 150 DEG C for example, removing
50 ° of data.It should be understood that the above-mentioned citing for error information is only illustrative examples, the adoptable error information of the present invention
It is without being limited thereto.
The pretreated actual operating data and early warning that warning module 504 is obtained based on data preprocessing module 503 are spread out
The amount of changing constructs Early-warning Model, specifically, Early-warning Model is constructed according to the rate of temperature change of pitch motor, further according to building
Early-warning Model carries out early warning to the blade cracking of wind power generating set.It is described in detail below with reference to Fig. 6 real according to the present invention
Apply the warning module 504 of example.
Fig. 6 shows the block diagram of the warning module of embodiment according to the present invention.
Referring to Fig. 6, warning module 504 includes that Early-warning Model establishes unit 601 and early warning judging unit 602.Specifically, in advance
Alert model foundation unit 601 constructs Early-warning Model according to the rate of temperature change of pitch motor, and early warning judging unit 602 is according to building
Early-warning Model seek the rate of temperature change accounting value of the wind generator set blade for early warning, and work as rate of temperature change accounting
Value triggers alarm when being more than or equal to the alarm value G for triggering blade cracking, wherein rate of temperature change accounting value is specific to meet
The quantity of the rate of temperature change of condition accounts for the ratio of the total amount of the rate of temperature change of pitch motor.According to an embodiment of the invention,
Early-warning Model establishes unit 601 and establishes assignment matrix corresponding with rate of temperature change, and the initial value of the element in assignment matrix is
0, then, it is determined that the data in rate of temperature change in the preset range, and by element assignment corresponding in assignment matrix
For the first value, adjacent and symbol in temperature change rate in preset range is opposite and/or is separated by and symbol is opposite
Any two data, and element value corresponding in assignment matrix is assigned a value of second value, meet in temperature change rate
Element value corresponding in assignment matrix is simultaneously assigned a value of third value by the data of the following conditions: adjacent and symbol it is opposite and/or
It is separated by and any two rate of temperature change that symbol is opposite, one in the absolute value of any two rate of temperature change is more than or equal to
Threshold alpha, another is less than threshold alpha, and the absolute value of the difference of any two rate of temperature change is greater than the data of threshold gamma, finally,
New assignment matrix is obtained by the assignment for carrying out the first value, second value and third value to the element in assignment matrix, to complete
The building of Early-warning Model.Wherein, specified conditions are α < < | Δ S | < β, α=0.11, β=0.657, threshold gamma=0.18.In advance
Alert judging unit 602 accounts for this newly according to the quantity that new assignment matrix calculates remaining first value corresponding with each pitch motor
The ratio value of assignment matrix total data line number to obtain rate of temperature change accounting value corresponding with each pitch motor, and judges
Alarm is triggered when rate of temperature change accounting value is more than or equal to the alarm value G of blade cracking.Here, alarm value is by close several
More units of damage and the actual operating data with the unit operated normally in the period are monitored what analysis obtained in year,
For example, alarm value G=0.01, then judge the rate of temperature change accounting value of the pitch motor of calculating, if rate of temperature change
Accounting value P is more than or equal to 0.01, then is judged as triggering early warning unit, triggers alarm, otherwise do not trigger alarm, early warning terminates.It answers
Understand, the above-mentioned citing for alarm value is only illustrative examples, and the adoptable alarm value of the present invention is without being limited thereto.
The method for early warning and system of a kind of wind generator set blade cracking of embodiment according to the present invention can be based on
The temperature of pitch motor constructs Early-warning Model, and then realizes the problem of big data early warning wind generator set blade cracks, and overcomes
The unconspicuous problem of temperature change of part MW class wind turbine turbines vane cracking and the detection of pitch motor sensor,
The real time on-line monitoring of blade operation and the work that gives warning in advance are realized, provides guarantee for the reliable and stable of blade.
It is may be recorded according to the method for the present invention including executing by the program instruction of computer implemented various operations
In computer-readable medium.Medium can also only include program instruction or include the data file combined with program instruction,
Data structure etc..The example of computer-readable medium includes magnetic medium (such as hard disk, floppy disk and tape)；Optical medium (such as
CD-ROM and DVD)；Magnet-optical medium (for example, CD)；And especially it is formulated for the hardware device for storing and executing program instructions
(for example, read-only memory (ROM), random access memory (RAM), flash memory etc.).Medium is also possible to include transmission regulation journey
The transmission medium (such as optical line or metal wire, waveguide etc.) of the carrier wave of the signal of sequence instruction, data structure etc..Program instruction
Example includes the machine code for example generated by compiler and the text comprising interpreter high-level code performed by computer can be used
Part.
Although the present invention, art technology has been shown and described referring to certain exemplary embodiments of the invention
Personnel will be understood that, can be into the case where not departing from the spirit and scope of the present invention being defined by the claims and their equivalents
Various changes on row various forms and details.
Claims (16)
1. a kind of method for early warning of wind generator set blade cracking, which is characterized in that the described method comprises the following steps:
Obtain the actual operating data of wind power generating set；
The derivative variable of early warning relevant to wind generator set blade cracking is calculated according to the actual operating data of acquisition；
Data prediction is carried out to the derivative variable of the early warning of the actual operating data of acquisition and calculating respectively；
Early-warning Model is constructed based on pretreated actual operating data and the derivative variable of early warning, and according to the Early-warning Model of building
It cracks to blade and carries out early warning.
2. the method as described in claim 1, which is characterized in that the actual operating data includes the acquisition of actual operating data
Time, the temperature of pitch motor and blade propeller pitch angle.
3. method according to claim 2, which is characterized in that described calculated according to the actual operating data of acquisition is sent out with wind-force
Electric turbines vane crack relevant early warning derivative variable the step of include:
Calculate the time difference of the acquisition time of the actual operating data, wherein when the time difference is adjacent in acquisition time
Between latter time subtract the previous time；
The temperature difference and pitch angular difference of pitch motor are accordingly calculated according to the time difference；
The rate of temperature change of pitch motor is calculated according to the temperature difference of pitch motor and time difference.
4. method as claimed in claim 3, which is characterized in that described respectively to the pre- of the actual operating data of acquisition and calculating
Warning the step of derivative variable carries out data prediction includes:
Remove the missing values and null value in the actual operating data；
Remove the error information in the derivative variable of the actual operating data and the early warning, wherein the error information includes
Data of the temperature of pitch motor more than or equal to 150 DEG C and pitch angular difference are greater than 50 ° of data.
5. method according to claim 2, which is characterized in that described to be spread out based on pretreated actual operating data and early warning
The amount of changing constructs Early-warning Model and includes: the step of carrying out early warning to blade cracking according to the Early-warning Model of building
Early-warning Model is constructed according to the rate of temperature change of pitch motor；
The rate of temperature change accounting value of the wind generator set blade for early warning is sought according to the Early-warning Model of building, and when temperature
Degree change rate accounting value triggers alarm when being more than or equal to the alarm value G for triggering blade cracking, wherein the rate of temperature change
Accounting value be meet the rate of temperature change of specified conditions quantity account for pitch motor rate of temperature change total amount ratio.
6. method as claimed in claim 5, which is characterized in that described to construct early warning mould according to the rate of temperature change of pitch motor
The step of type includes:
Assignment matrix corresponding with rate of temperature change is established, the initial value of the element in the assignment matrix is 0；
Data in temperature change rate in the preset range, and element corresponding in assignment matrix is assigned a value of the
One value；
Adjacent and symbol in temperature change rate in the preset range is opposite and/or is separated by and what symbol was opposite appoints
It anticipates two data, and element value corresponding in assignment matrix is assigned a value of second value；
Meet the data of the following conditions in temperature change rate and element value corresponding in assignment matrix is assigned a value of
Three values: adjacent and symbol is opposite and/or is separated by and any two rate of temperature change that symbol is opposite, any two temperature become
One in the absolute value of rate is more than or equal to threshold alpha, another is less than threshold alpha, and any two rate of temperature change it
Absolute value of the difference is greater than the data of threshold gamma；
New assignment matrix is obtained by the assignment for carrying out the first value, second value and third value to the element in assignment matrix, with
The building of Early-warning Model is completed,
Wherein, the preset range is α < < | Δ S < β, Δ S are rate of temperature change, threshold alpha=0.11, the threshold value beta
=0.657, threshold gamma=0.18, the alarm value G=0.01.
7. method as claimed in claim 6, which is characterized in that the Early-warning Model according to building seeks the wind for early warning
The step of rate of temperature change accounting value of power generator group blade includes:
The new assignment matrix is accounted for according to the quantity that new assignment matrix calculates remaining first value corresponding with each pitch motor
The ratio value of total data line number, to obtain rate of temperature change accounting value corresponding with each pitch motor.
8. a kind of early warning system of wind generator set blade cracking, which is characterized in that the system comprises:
Actual operating data obtains module, is configured as obtaining the actual operating data of wind power generating set；
Early warning derives variable computing module, is configured as according to the calculating of the actual operating data of acquisition and wind generator set blade
The derivative variable of relevant early warning of cracking；
Data preprocessing module is configured to count the actual operating data of acquisition and the derivative variable of the early warning of calculating
Data preprocess；
Warning module is configured as based on pretreated actual operating data and the derivative variable building Early-warning Model of early warning, and
It is cracked according to the Early-warning Model of building to blade and carries out early warning.
9. system as claimed in claim 8, which is characterized in that the actual operating data is divided into wind power generating set
Operate normally data and fault moment data, the temperature of acquisition time, pitch motor including actual operating data and blade
Propeller pitch angle.
10. system as claimed in claim 9, which is characterized in that the derivative variable computing module of the early warning is configured as:
Calculate the time difference of the acquisition time of the actual operating data, wherein when the time difference is adjacent in acquisition time
Between latter time subtract the previous time；
The temperature difference and pitch angular difference of pitch motor are accordingly calculated according to the time difference；
The rate of temperature change of pitch motor is calculated according to the temperature difference of pitch motor and time difference.
11. system as claimed in claim 10, which is characterized in that the data preprocessing module is configured as:
Remove the missing values and null value in the actual operating data；
Remove the error information in the derivative variable of the actual operating data and the early warning, wherein the error information includes
Data of the temperature of pitch motor more than or equal to 150 DEG C and pitch angular difference are greater than 50 ° of data.
12. system as claimed in claim 10, which is characterized in that the warning module is configured as:
Early-warning Model establishes unit, constructs Early-warning Model according to the rate of temperature change of pitch motor；
Early warning judging unit seeks the rate of temperature change of the wind generator set blade for early warning according to the Early-warning Model of building
Accounting value, and alarm is triggered when rate of temperature change accounting value is more than or equal to the alarm value G for triggering blade cracking, wherein
The rate of temperature change accounting value is to meet the quantity of the rate of temperature change of specified conditions to account for the rate of temperature change of pitch motor
The ratio of total amount.
13. system as claimed in claim 12, which is characterized in that the Early-warning Model is established unit and is configured as:
Assignment matrix corresponding with rate of temperature change is established, the initial value of the element in the assignment matrix is 0；
Data in temperature change rate in the preset range, and element corresponding in assignment matrix is assigned a value of the
One value；
Adjacent and symbol in temperature change rate in the preset range is opposite and/or is separated by and what symbol was opposite appoints
It anticipates two data, and element value corresponding in assignment matrix is assigned a value of second value；
Meet the data of the following conditions in temperature change rate and element value corresponding in assignment matrix is assigned a value of
Three values: adjacent and symbol is opposite and/or is separated by and any two rate of temperature change that symbol is opposite, any two temperature become
One in the absolute value of rate is more than or equal to threshold alpha, another is less than threshold alpha, and any two rate of temperature change it
Absolute value of the difference is greater than the data of threshold gamma；
New assignment matrix is obtained by the assignment for carrying out the first value, second value and third value to the element in assignment matrix, with
The building of Early-warning Model is completed,
Wherein, the specified conditions are α < < | Δ S | < β, Δ S are rate of temperature change, α=0.11, β=0.657, the threshold
Value γ=0.18, the alarm value G=0.01.
14. method as claimed in claim 12, which is characterized in that the early warning judging unit is also configured to
The new assignment matrix is accounted for according to the quantity that new assignment matrix calculates remaining first value corresponding with each pitch motor
The ratio value of total data line number, to obtain rate of temperature change accounting value corresponding with each pitch motor.
15. a kind of computer readable storage medium, is stored with computer program, which is characterized in that the computer program is located
When managing device operation, the processor perform claim requires method described in any one of 1-7.
16. a kind of computer equipment, the memory including processor and storage computer program, which is characterized in that the calculating
When machine program is run by processor, the processor perform claim requires method described in any one of 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810393681.0A CN110414022B (en) | 2018-04-27 | 2018-04-27 | Early warning method and system for cracking of wind generating set blade |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810393681.0A CN110414022B (en) | 2018-04-27 | 2018-04-27 | Early warning method and system for cracking of wind generating set blade |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110414022A true CN110414022A (en) | 2019-11-05 |
| CN110414022B CN110414022B (en) | 2023-05-09 |
Family
ID=68346927
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810393681.0A Active CN110414022B (en) | 2018-04-27 | 2018-04-27 | Early warning method and system for cracking of wind generating set blade |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110414022B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113357097A (en) * | 2020-03-02 | 2021-09-07 | 北京金风科创风电设备有限公司 | Blade clamping detection method and device of wind generating set |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102721924A (en) * | 2012-06-26 | 2012-10-10 | 新疆金风科技股份有限公司 | Fault early warning method of wind generating set |
| CN104807644A (en) * | 2015-04-14 | 2015-07-29 | 北京中恒博瑞数字电力科技有限公司 | Fault early warning method and system for wind generation set variable-pitch system |
| US20150337802A1 (en) * | 2014-05-26 | 2015-11-26 | General Electric Company | System and method for pitch fault detection |
| CN107449102A (en) * | 2017-06-28 | 2017-12-08 | 深圳达实智能股份有限公司 | Hospital's fan coil self-diagnosis of trouble method, apparatus and readable storage medium storing program for executing |
-
2018
- 2018-04-27 CN CN201810393681.0A patent/CN110414022B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102721924A (en) * | 2012-06-26 | 2012-10-10 | 新疆金风科技股份有限公司 | Fault early warning method of wind generating set |
| US20150337802A1 (en) * | 2014-05-26 | 2015-11-26 | General Electric Company | System and method for pitch fault detection |
| CN104807644A (en) * | 2015-04-14 | 2015-07-29 | 北京中恒博瑞数字电力科技有限公司 | Fault early warning method and system for wind generation set variable-pitch system |
| CN107449102A (en) * | 2017-06-28 | 2017-12-08 | 深圳达实智能股份有限公司 | Hospital's fan coil self-diagnosis of trouble method, apparatus and readable storage medium storing program for executing |
Non-Patent Citations (1)
| Title |
|---|
| 仲政: "固体热-化-力耦合连续介质理论的研究现状与展望", 《力学季刊》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113357097A (en) * | 2020-03-02 | 2021-09-07 | 北京金风科创风电设备有限公司 | Blade clamping detection method and device of wind generating set |
| CN113357097B (en) * | 2020-03-02 | 2024-01-26 | 北京金风科创风电设备有限公司 | Blade clamping detection method and device for wind generating set |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110414022B (en) | 2023-05-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| ES2973349T3 (en) | Scalable system and engine to predict wind turbine failures | |
| CN102022264B (en) | System and method for wind turbine health management | |
| CN111597682B (en) | Method for predicting remaining life of bearing of gearbox of wind turbine | |
| US10902163B2 (en) | Simulation method and system | |
| Vera-Tudela et al. | Analysing wind turbine fatigue load prediction: The impact of wind farm flow conditions | |
| CN109543210A (en) | A kind of Wind turbines failure prediction system based on machine learning algorithm platform | |
| US11761427B2 (en) | Method and system for building prescriptive analytics to prevent wind turbine failures | |
| CN111950505B (en) | SSA-AANN wind driven generator sensor state evaluation method | |
| CN106779200A (en) | Based on the Wind turbines trend prediction method for carrying out similarity in the historical data | |
| KR20160073945A (en) | System and method for managing wind plant | |
| JP2013222423A (en) | Method and system for predicting power generation capacity, and method and system for managing health of wind power generation facilities | |
| CN106870298A (en) | Blade root bolt fracture fault detection method based on machine learning | |
| CN110362045A (en) | A kind of marine double-fed fan motor unit fault distinguishing method considering maritime meteorology factor | |
| CN112832960A (en) | Fan blade icing detection method based on deep learning and storage medium | |
| CN114423940A (en) | Method for estimating remaining useful life of a component of an operational wind turbine | |
| CN110412966A (en) | The method and apparatus for monitoring pitch motor temperature anomaly | |
| CN109595130A (en) | A kind of fan blade icing failure prediction method and system | |
| CN116306139A (en) | Intelligent monitoring method and system for service life of wind turbine blade | |
| CN109236589B (en) | It is a kind of for assessing the method and device of fan blade deicing capital project | |
| CN103092914A (en) | Expert system knowledge acquisition method based on Weka software | |
| CN110414022A (en) | A kind of method for early warning and system of wind generator set blade cracking | |
| CN114463143A (en) | Method for enhancing SCADA fault data of offshore doubly-fed wind generator | |
| Chen et al. | A novel multi-task learning method with attention mechanism for wind turbine blades imbalance fault diagnosis | |
| CN116644968A (en) | Wind turbine generator sub-component state analysis method based on state sensing space | |
| CN112202924A (en) | High-power wind turbine generator system bolt intellectual detection system that excels in based on thing networking |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 830026 No. 107, Shanghai Road, Urumqi economic and Technological Development Zone, the Xinjiang Uygur Autonomous Region Patentee after: Jinfeng Technology Co.,Ltd. Address before: 830026 No. 107, Shanghai Road, Urumqi economic and Technological Development Zone, the Xinjiang Uygur Autonomous Region Patentee before: XINJIANG GOLDWIND SCIENCE & TECHNOLOGY Co.,Ltd. |